The main SaaS pricing models are flat rate, per-seat, tiered (packaged plans), usage-based, and freemium, and most companies mix two or more.
Real price tests point the same way: pricing that scales with the value a customer gets, sold in a few clear packages, tends to earn more revenue per visitor than cheap or purely per-unit pricing, even when fewer people sign up.
This page covers the models, how to pick a strategy, what value-based pricing looks like in practice, and the monthly vs annual billing question, all backed by price tests from our database.
SaaS pricing models compared
Stripe's billing docs give a useful baseline set of recurring pricing models: flat rate, per-seat, tiered, and usage-based.
Freemium and free trials sit on top of these as ways to get people in the door.
| Model | How the price is set | Works well when | Watch out for |
|---|---|---|---|
| Flat rate | One price for the product, or one price per named plan | Customers are similar and value doesn't vary much | Big customers pay the same as small ones |
| Per-seat | Price times number of users | Value grows with each person using it (team tools) | Teams share logins to save money |
| Tiered / packaged | Good-better-best plans split by features, limits or support | You can segment buyers by need | Too many plans or unclear differences |
| Usage-based | Charged per unit used (API calls, messages, storage) | Usage tracks the value customers get | Unpredictable bills; hard to forecast revenue |
| Freemium | Free plan with limits, paid plans above it | Free users spread the product or cost little to serve | Free plan good enough that few upgrade |
Stripe also splits tiered pricing into two kinds.
With volume pricing, every unit is charged at the rate of the tier you land in.
With graduated pricing, each unit is charged at the rate of the tier it falls in, and the amounts are added up.
The clearest model test in our database is Server Density's switch away from per-unit pricing.
In a VWO test, packaged tiers starting at US$99 a month replaced per-server pricing and revenue rose 114% ($394 to $833), while free sign-ups fell 24.96% (6.92% to 5.19%).
The paid conversion counts were small (20 vs 15), so treat it as strong direction, not a precise lift.
Patrick McKenzie's write-up of the same experiment reports it as about +100% revenue: trial sign-ups fell from 150 to 113, while revenue added per visitor went from about $0.19 to $0.41. VWO showed 93% confidence on revenue.
McKenzie's argument was that "pricing scaling linearly when customer value scales exponentially is a poor decision."
How to build a SaaS pricing strategy
A pricing strategy is the set of choices behind the numbers: who you price for, what you charge for, how many packages you offer, and how you'll judge a change.
Price level gets the attention, but packaging and the metric you charge on often matter more.
- Pick the right success metric before you test. Sign-ups and revenue often move in opposite directions. Use revenue per visitor or revenue per user, not conversion rate alone.
- Segment by need, not by unit count. Plans split by team size, support level, or key limits let serious buyers self-select into higher tiers.
- Keep the plan count small. Three or four plans with one highlighted is the common pattern; see SaaS pricing page tests for layout evidence.
- Decide the entry offer. Free plan, free trial, or paid only. This choice can move paid conversion more than the price itself.
- Test bold changes. A 5% price change rarely shows up in a test with SaaS traffic levels. Packaging changes and big price steps do.
The entry offer point has strong evidence. Acuity Scheduling sent its main sign-up button to a 14-day trial of its $10/month Professional plan instead of a free-forever account, and paid Professional sign-ups rose 268.14% at a 99.99% chance to beat the original.
That test also asked for card details up front, so it changed the offer, not just the page.
Our SaaS free trial guide and freemium vs free trial page go deeper.
The success metric point matters just as much.
In a five-way Boston Globe test written up by MarketingExperiments, a $3.99 regular weekly rate after a $0.99 intro produced the most total revenue, 163% more than the $4.99 version.
But the $0.99 control won on subscriptions and had the lowest cancellation rate, and $4.99 had the highest revenue per subscriber.
Each of those could be called the winner, depending on the KPI.
Value-based pricing for SaaS
Value-based pricing means setting prices from what the product is worth to the customer, not from your costs or competitor prices.
In SaaS, that usually comes down to choosing a value metric (the thing you charge on) that grows as the customer gets more out of the product.
The tests suggest three practical lessons.
Charge on the limit customers care about
Powtoon tested storage limits on its pricing page.
The winning version, run by Wingify Services, showed the Business plan at 10 GB instead of "unlimited" next to Pro at 2 GB, and revenue rose 27.9%.
A concrete, larger limit made the step up easier to judge than an open-ended promise.
Higher prices often lose buyers but win revenue
BuyBye, a savings app, raised its yearly price 50% with the same design: conversions fell 19% and proceeds per user rose 25%.
Mojo, a video editing app, saw only a minimal drop in free-to-paid conversion after a 50% yearly price rise in markets like the USA and Germany, though it published no exact figures.
These are mobile subscription tests, but the math is the same for web SaaS.
Cheaper isn't always worse, and isn't always better
Value-based pricing doesn't mean "always charge more." In an old four-day MarketingExperiments test, $10 a month beat $12.50 and $14.95: 156 orders and $1,560 revenue, against $1,175 and $1,106. That's 33% more revenue than the next best price, though no significance figures were given.
Meanwhile PriceCharting found that showing "Starts at $4" beat "Starts at $2" and "Starts at $1" next to a button, with 91% significance and no clear winner.
A price that looks too low can signal low value.
For the psychology behind price display, see anchoring and the decoy effect.
Treat both as ideas to test, not guaranteed lifts.
Monthly vs annual billing
Offering both monthly and annual billing, with a discount for paying yearly, is the default for most SaaS companies.
The tests suggest that giving people a term choice helps, and that how you frame the annual price matters.
Taking the choice away can hurt badly. In a MarketingExperiments test, removing the subscription-length dropdown so visitors were locked into monthly cut conversion by about 40%.
The business was anonymized and may have been a physical delivery subscription, but the lesson about term choice carries over.
Adding a longer term can pay off too.
Meebox, a Danish hosting company, offered 20-40% off for a 2-year commitment and saw revenue rise 121.56%, average order value rise 46.24%, and conversions rise 51.85%, at a 98% chance to beat the original.
Framing is the next lever. Mojo tested showing its annual plan as a per-month price, billed annually.
New revenue per paywall impression rose 45% in Brazil, 26% in Mexico and 8% in the USA.
Preu AI added an annual plan next to its monthly-only offer, priced at 60% of the monthly price, and proceeds per user rose 80%, with trial conversion up 22%.
Not every billing display change wins. Whitepages' redesign with side-by-side plans and a monthly/annual toggle lifted conversion 23% and monthly memberships 31%, but it bundled several layout changes and ran for only one week.
Powtoon's test of a new monthly vs annual price display lost, and no numbers were published.
| Billing choice | Good for | Risk |
|---|---|---|
| Monthly only | Low commitment for new buyers | Higher churn exposure; less cash up front |
| Monthly + annual (discounted) | Lets buyers self-select; more cash up front | Discount too deep cuts revenue per customer |
| Annual shown as monthly price | Makes the yearly plan look easier to afford | Must be clear it's billed annually, or refunds and complaints rise |
| Annual only | Simple; best cash flow | Loses buyers who won't commit for a year |
For more app-side billing tests, see subscription pricing for apps.
MarketingExperimentsMarketingExperimentsHow to test SaaS pricing without fooling yourself
Price tests are harder than button tests because the outcome you care about (revenue, retention) happens later and with fewer events. A few rules from the tests above:
- Measure revenue, not just sign-ups. Server Density's winning version lost about a quarter of its trial sign-ups and still roughly doubled revenue.
- Plan for small numbers. Paid conversions are rare, so you need far more traffic than you think. Check your sample size before you start.
- Watch what happens after the sale. A cheap intro price can win on sign-ups and lose on cancellations, as the Boston Globe test showed.
- Test big steps. Packaging, the value metric, and 30-50% price moves are visible in a test. Small tweaks usually aren't.
Our price testing guide covers the setup in detail, and the SaaS hub collects all our SaaS tests.